Magnetic and Structural Properties of the Iron Oxychalcogenides La$_{2}$O$_{2}$Fe$_{2}$O$M_{2}$ ($M$= S, Se)

We present the results of structural and magnetic phase comparisons of the iron oxychalcogenides La$_{2}$O$_{2}$Fe$_{2}$O$M$$_{2}$ ($M$ = S, Se). Elastic neutron scattering reveals that $M$ = S and Se have similar nuclear structures at room and low temperatures. We find that both materials obtain antiferromagnetic ordering at a Neel temperature $T_{N}$ 90.1 $\pm$ 0.16 K and 107.2 $\pm$ 0.06 K for $M$= Se and S, respectively. The magnetic arrangements of $M$ = S, Se are obtained through Rietveld refinement. We find the order parameter exponent $\beta$ to be 0.129 $\pm$ 0.006 for $M$ = Se and 0.133 $\pm$ 0.007 for $M$ = S. Each of these values is near the Ising symmetry value of 1/8. This suggests that although lattice and electronic structural modifications result from chalcogen exchange, the nature of the magnetic interactions is similar in these materials.


I. INTRODUCTION
The discovery of superconductivity in iron pnictide (FeP n) compounds has generated considerable interest because these materials seem to be the only current alternative to the cuprates for comparably high transition temperatures to the superconducting state. As with the cuprates, superconductivity in the iron pnictides appears by doping electrons or holes into a magnetically ordered parent compound. An important question is whether the iron pnictide parent compounds are on the verge of a metal insulator transition. [1,2] In order to establish the so-called incipient Mott scenario [3] in iron pnictides, it is important to identify the Mott insulating portion of phase diagram of these materials. One way to drive an Fe pnictide into the Mott insulating phase is by reducing the electron kinetic energy t and increasing the electron correlation interaction U .
The iron oxychalcogenides La 2 O 2 Fe 2 OM 2 M = (S, Se) provide a case study in this approach. These particular iron oxychalcogenides are parent compounds that have a composition such that the nominal valence of Fe is 2+ and contain an Fe square lattice, which is similar to, but expanded relative to the iron pnictides. La 2 O 2 Fe 2 O(S, Se) 2 were first reported to have insulating properties by Mayer. [4] In addition, the crystal structure contains tetragonally ordered, FeP n-like Fe planes for which chalcogens alternate above and below the iron atoms. Oxygen atoms are contained in the RE layers remniscent of the charge resevoirs of high-T c cuprates. [4,5] Specific attention has been given to iron oxychalcogenides [6][7][8][9] by investigators seeking to discover new iron-based materials in which high-T c superconductivity might be obtained by doping the Mott insulating state. [5,10,11] In addition, the electronic behavior of La 2 O 2 Fe 2 O(S, Se) 2 has been investigated. Zhu et.
al., using local density to dynamical mean-field theory (LDA + DMFT), predicted [5] Mott insulating behavior and band narrowing of La 2 O 2 Fe 2 O(S, Se) 2 . Bulk transport and magnetic measurements, in addition to resonant inelastic x-ray scattering (RIXS) and soft x-ray absorption spectroscopy [12], provided experimental evidence that the systems are indeed Mott insulators. [5] Band narrowing within La 2 O 2 Fe 2 O(S, Se) 2 has been proposed to lead to a Mott insulating state as well as enhanced electron correlation effects. While exhibiting strongly correlated Mott insulating behavior, La 2 O 2 Fe 2 O(S, Se) 2 may offer tunability of their electronic properties near a metal-insulator transition (MIT). In Mott insulators, it has been observed that sizable electronic correlations drives new physical effects upon 3 doping (electron or hole) and other external perturbations. [13] They can induce a range of interesting behavior including a pseudogap regime as in the case of Na 2 Fe 2 OSe 2 [8] [14] or orbital-selective incoherent states that naturally yield co-existent insulating and bad metallic states as in cuprates or iron-pnictides. [15][16][17] Furthermore, iron oxychalcogenides can be tuned by transition-metal or chalcogen substitution [5], to produce novel electronic and magnetic phases at low temperature. The substitution of S and Se has been shown to alter the character of electronic partial density of states in this material class. [5,12,18] Even the presence of superconductivity in the iron oxychalcogenides has been of interest such that the effects of F-doping in La 2 O 3−x F x Fe 2 Se 2 and the substitution of Mn for Fe in La 2 O 2 Fe 1−x Mn x Se 2 have been investigated; however, no HTSC was observed. [11] [19] [20] In addition to interesting electronic properties, studies of the magnetic behavior of iron oxychalcogenides have been pursued. A 2 F 2 TM 2 O(M ) 2 where A = (Ba, Sr) have been the subject of recent studies [5,[19][20][21][22][23][24][25] which showed that these materials order antiferromagnetically. Further, Ba 2 F 2 Fe 2 OSe 2 was proposed to be an example of a compound with a frustrated AFM checkerboard spin lattice. [22] Other oxychalcogenides exhibit the onset of AFM ordering 2D short-range magnetic correlation well above T N . [ In this work, we study and compare the structural and magnetic properties of the iron oxychalcogenides La 2 O 2 Fe 2 OS 2 and La 2 O 2 Fe 2 OSe 2 using neutron powder diffraction (NPD).
Our focus is on powder materials since single-crystalline samples remain difficult to produce.
We measure the nuclear and magnetic Bragg scattering intensity as a function of temperature and we examine the structural distinctions between the two parent compounds at room and low temperatures. Section II provides the experimental details of the neutron and transport measurements. Section III gives the results of structural and magnetic diffraction of M = (S, Se). In addition, we discuss the magnetic structure and the magnetic order parameter behavior which reveal Ising symmetry in both materials. We provide a discussion of our findings within the context of specific magnetic exchange interactions and relative to other oxychalcogenides reports in the literature. Finally, our results are discussed in light of some theoretical findings that have been reported on La 2 O 2 Fe 2 O(S, Se) 2 systems.

II. EXPERIMENT
The samples studied here La 2 O 2 Fe 2 O(S, Se) 2 have nominal compositions and were prepared by a conventional solid-state reaction method using high purity La 2 O 3 , S, Se and Fe powders as starting materials. The powders were mixed in the stoichiometric ratios and carefully ground. Subsequently, the powders were pressed into pellets and then heated in an evacuated quartz tube at 1030 • C for 3 days; this process was repeated three times. The samples were confirmed to be of a single phase by the laboratory X-ray powder diffraction measurements. [4] Neutron powder diffraction (NPD) experiments were performed using the C2 highresolution diffractometer at the NRU reactor at Chalk River Canadian Nuclear Labs. Room Resistivity versus temperature data for M = S and Se have been published in Refs. [5] and [12]. M is defined as the magnetization per unit volume and H is the applied magnetic field (1 T in our measurement setup). The magnetic susceptibility dM/dH as a function of temperature is shown in Fig. 1  Room temperature data was collected for crystal structure refinement in order to avoid magnetic Bragg peak contributions in analyzing the structural details of these materials. A direct comparison of the Rietveld refinement parameters of M = (S, Se) is given in Table I. Fe unit cell volume of M = Se is larger than that of M = S. This is reasonable given that the ionic radius of sulfur (100 pm) is smaller than selenium (115 pm) and we find that d F e−F e values reported in LaFeOAs, the interatomic distances we obtained (in Table II) are larger by 1.2 % and nearly 1.1% for M = Se and S, respectively. Bond angles and atomic distances, extracted from Rietveld refinement parameters, are tabulated for comparison of M = S and Se (see Table II). Rietveld analysis yielded measurements for the bond angles subtended by Se-Fe-Se and S-Fe-S defined as θ 1 and θ 2 , respectively. These bonds define the distortion of the Fe(S, Se) 4 squares contained in the Fe 2 OS 4 and Fe 2 OSe 4 octahedra (c.f . Figure 3(a)), respectively. For completeness Table III presents the atomic position of The atomic structure (see Fig. 3 (8) 4e  located, above or below, the center of the Fe-O plaquettes; therefore, the Fe-M layers are not flat. These D 2h point symmetry octahedra are face sharing (c.f . Fig. 3 (a)) such that the shared face is intersected by the Fe-Fe nearest-neighbor line-of-sight. [21] Here we list the angles θ 1 and θ 2 (see Fig. 3  was conducted by extracting atomic position deviations from radial distribution function data. [33] No local structure change from the low-temperature I /4mmm symmetry was observed in these experiments. Both Fuwa [29] and Free [28] suggested that the absence of   [5], [28] and [37].    Fig. 6(b). The 2-k structure is made up of 2 orthogonal stripes within the Fe 2 OM 2 layers. While the AFM3 configuration provided good fits to our M = S and Se NPD data, as noted [37], neutron powder diffraction can not  To determine the thermal dependence of the magnetic ordering behavior of M = (S, Se), the intensity of the magnetic Bragg peak Q = (-103) was measured over a temperature range of 300 K to 4 K in each material. The peak intensity can be used as a measure of the magnetic order parameter squared φ 2 . The square of the magnetic order parameters of this produces different magnitudes of distortion within the octahedra. This distortion is expected to be related to the presence of the relatively high extent of electron correlation compared to the iron pnictides. In addition, the distorted octahedra can diminish magnetoelastic coupling by precluding orbital ordering that is necessary to establish a link between the magnetic phase transition and a structural phase transformation. [32] We did not observe structural phase transitions in the materials. Nor did we see evidence of a nematic phase similar to that which exists in the iron pnictides. However, observing only a magnetic phase transition from the high-temperature paramagnetic phase to a lower temperature AF phase, we used group theory and magnetic refinement methods to determine the magnetic We discussed models of frustrated magnetism and their relevance to metallic and insulating behavior iron oxychalcogenides.

V. AKNOWLEDGEMENTS
We are grateful to the technical staff at CNBC for excellent support.